Resource recommendation in social annotation systems: A linear-weighted hybrid approach

Gemmell, Jonathan; Schimoler, Thomas; Mobasher, Bamshad; Burke, Robin

doi:10.1016/j.jcss.2011.10.006

Cited by 40 publications

(29 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The KNN method is popularly used in collaborative filtering recommendation systems [7,10,17,30]. In this study, the traditional KNN method is revised and adapted to the gift giving model.…”

Section: Data Mining Techniquesmentioning

confidence: 99%

Mining the Gift Receiver's Mind

Tu¹,

Fu²

2017

IJDKP

View full text Add to dashboard Cite

Choosing an appropriate gift is difficult because the purpose of gift giving is to arouse affection in the receiver, not the giver, and too many variables that influence the results. Utilizing 600 samples and a hybrid method combining the decision tree and K-nearest neighbor approaches, this study builds a DTKNN two-stepped recommendation system which achieves a precision rate higher than 80%. The contribution of this research is to propose a new data mining technique to solve the problem of a recommendation system for altruistic gift selection which allows the receiver to perceive the affection desired by the giver. K EY W ORDS:Recommendation system, Gift giving, Decision tree, K-nearest neighbor.

show abstract

“…The KNN method is popularly used in collaborative filtering recommendation systems [7,10,17,30]. In this study, the traditional KNN method is revised and adapted to the gift giving model.…”

Section: Data Mining Techniquesmentioning

confidence: 99%

Mining the Gift Receiver's Mind

Tu¹,

Fu²

2017

IJDKP

View full text Add to dashboard Cite

show abstract

“…However, the accuracy of the predictions is affected mostly by the collaborative filtering prediction, and occasionally the weight value is not precise. Jonathan G. et al [22] come up with a linear-weighted hybrid framework for making recommendations in social annotation systems, and experiments were conducted using six real-world datasets to prove that the recommender system was more effective and flexible. Gong et al [23] extracted information from social networks related to users, which can be integrated into a collaborative filter to improve the performance of the system.…”

Section: Related Workmentioning

confidence: 99%

An Efficient Recommendation Filter Model on Smart Home Big Data Analytics for Enhanced Living Environments

Chen

Xie

Shu

et al. 2016

Sensors

View full text Add to dashboard Cite

With the rapid growth of wireless sensor applications, the user interfaces and configurations of smart homes have become so complicated and inflexible that users usually have to spend a great amount of time studying them and adapting to their expected operation. In order to improve user experience, a weighted hybrid recommender system based on a Kalman Filter model is proposed to predict what users might want to do next, especially when users are located in a smart home with an enhanced living environment. Specifically, a weight hybridization method was introduced, which combines contextual collaborative filter and the contextual content-based recommendations. This method inherits the advantages of the optimum regression and the stability features of the proposed adaptive Kalman Filter model, and it can predict and revise the weight of each system component dynamically. Experimental results show that the hybrid recommender system can optimize the distribution of weights of each component, and achieve more reasonable recall and precision rates.

show abstract

“…The WHyLDR model has been shown to meet both of these needs in the area of social tagging systems, performing basic recommendation tasks with accuracy surpassing that of single-purpose model-based techniques such as tensor factorization, and also supporting a wide variety of recommendation tasks [8].…”

Section: Recsysmentioning

confidence: 99%

“…Our approach, called the Weighted Hybrid of Low-Dimensional Recommenders (WHyLDR) is designed to support the flexible creation and rapid deployment of a wide variety of recommenders in a heterogeneous environment. We have demonstrated its effectiveness in prior work focusing on social tagging systems [8,7,6,3].…”

Section: Introductionmentioning

confidence: 99%

Weighted hybrid recommendation for heterogeneous networks

Vahedian

2014

Proceedings of the 8th ACM Conference on Recommender Systems

View full text Add to dashboard Cite

Social media sites accumulate a wide variety of information about users: likes and ratings, friend and follower links, annotations, posts, media uploads, just to name a few. Key challenges for recommender systems research are (a) to synthesize of all of this data into an integrated recommendation model and (b) to support a wide variety of recommendation types simultaneously (items, friends, tags, etc.) One approach that has been explored in recent research is to view this multi-faceted data as a heterogeneous network and use network-based methods of generating recommendations. However, most such approaches involve computationallyintensive model generation resulting in a single-purpose recommender system. Our approach is to create a componentbased hybrid model whose components can be reused for multiple recommendation tasks. In this paper, we show how this model can be applied to heterogeneous networks.

show abstract

Resource recommendation in social annotation systems: A linear-weighted hybrid approach

Cited by 40 publications

References 8 publications

Mining the Gift Receiver's Mind

Mining the Gift Receiver's Mind

An Efficient Recommendation Filter Model on Smart Home Big Data Analytics for Enhanced Living Environments

Weighted hybrid recommendation for heterogeneous networks

Contact Info

Product

Resources

About